[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <503339cf-7e54-0888-1767-c8ac87ce2130@linux.intel.com>
Date: Mon, 2 Apr 2018 13:41:57 -0700
From: Dave Hansen <dave.hansen@...ux.intel.com>
To: Linus Torvalds <torvalds@...ux-foundation.org>
Cc: Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
linux-mm <linux-mm@...ck.org>,
Andrea Arcangeli <aarcange@...hat.com>,
Andrew Lutomirski <luto@...nel.org>,
Kees Cook <keescook@...gle.com>,
Hugh Dickins <hughd@...gle.com>,
Jürgen Groß <jgross@...e.com>,
the arch/x86 maintainers <x86@...nel.org>, namit@...are.com
Subject: Re: [PATCH 09/11] x86/pti: enable global pages for shared areas
On 04/02/2018 10:56 AM, Linus Torvalds wrote:
> On Mon, Apr 2, 2018 at 10:27 AM, Dave Hansen
> <dave.hansen@...ux.intel.com> wrote:
>> + /*
>> + * The cpu_entry_area is shared between the user and kernel
>> + * page tables. All of its ptes can safely be global.
>> + */
>> + if (boot_cpu_has(X86_FEATURE_PGE))
>> + pte = pte_set_flags(pte, _PAGE_GLOBAL);
> So this is where the quesion of "why is this conditional" is valid.
>
> We could set _PAGE_GLOBAL unconditionally, not bothering with testing
> X86_FEATURE_PGE.
I think we should just keep the check for now. Before this patch set,
on !X86_FEATURE_PGE systems, we cleared _PAGE_GLOBAL in virtually all
places due to masking via __supported_pte_mask.
It is rather harmless either way, but being _consistent_ (by keeping the
check) with all of our PTEs is nice.
Powered by blists - more mailing lists